A glimpse into the AI-powered Zoom Contact Center add-on that’s designed to improve a contact center’s operations, agent performance, and consumer satisfaction.
Zoom’s Quality Management is a contact center add-on designed to improve a contact center’s operations, agent performance, and consumer satisfaction
Providing a positive experience is paramount to a contact center’s operations. To help simplify this process, Zoom's Quality Management is a contact center add-on specifically designed to enhance operational and agent performance and improve consumer satisfaction. Quality Management accomplishes this through various tools and features that monitor and analyze consumer interactions, evaluate agent performance, offer actionable insights, and proactively identify areas for improvement.
Quality Management is powered through transcription and recording analysis
At its core, Quality Management uses artificial intelligence (AI) and machine learning (ML) algorithms to analyze transcripts of contact center engagements, and provides insights and analyses that contact center teams can use to provide better consumer experiences.
Within its framework, Quality Management includes features that help to ensure organizational compliance and provide insights into consumer sentiment
Because Quality Management requires the recording of contact center engagements for analysis, accounts can review entire conversations to help verify organizational compliance. These recordings can be coupled with Quality Management’s built-in tools like scorecards and evaluations, that assist managers and supervisors with assessing agent interactions with consumers, helping to ensure adherence to established protocols and procedures.
Additionally, Quality Management is designed to discern consumer sentiment through transcript analysis based on word choice and phrasing. For instance, moments of explicit language, requests to speak with a manager, or expressions of frustration will be reflected in the sentiment score of the conversation. This allows for proactive measures to address consumer dissatisfaction before it escalates to critical levels.
Zoom’s Quality Management includes speech analytics at no additional cost
Although some contact center solutions offer quality management features separate from speech analytics, Zoom’s Quality Management includes both within the same product at no additional cost to the customer.
Quality Management focuses on specific interactions, not total engagements
A contact center engagement is defined as the specific instance of a customer's contact center experience, such as a customer calling in for help with a problem. Within each engagement, there are one or more interactions, which are defined as unique contact points between a customer and an agent. Zoom Quality Management focuses on these individual interactions.
For example, in one engagement (e.g.., phone call), a customer might have three interactions: one with tier-one support, one with tier-two support, and one with a manager. If the customer hangs up and calls again, a new engagement begins, which may also include multiple interactions.
Zoom Quality Management empowers supervisors to review each specific interaction within the overall engagement, allowing them to focus on individual contact points rather than the entire engagement. However, supervisors can still easily see and navigate between all interactions associated with a common engagement.
Quality Management currently supports the voice and video contact channels
Quality Management currently supports functionality with the voice and video contact channels. In future releases, Quality Management will expand to support additional channels.
Quality Management is exclusively available with Zoom Contact Center
Quality Management is currently available exclusively as an add-on for Zoom Contact Center. Quality Management cannot be integrated with other third-party contact center solutions at this time.
Quality Management currently supports English and 17 additional languages in beta for conversational Insights and transcription
Quality Management currently supports English and 17 additional languages in beta for transcription and conversational Insights, including:
|
|
|
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
Quality Management is a part of Zoom’s Workforce Engagement Management Suite
Quality Management is a component of Zoom's Workforce Engagement Management (WEM) suite, which comprises add-on solutions designed to enhance contact center capabilities. When used in conjunction with Workforce Management — another WEM suite product — contact center managers can simplify their contact center management. Features of Workforce Management include agent forecasting, scheduling, shift management, time-off management, and more. To learn more about the Workforce Engagement Management suite, reach out to your Zoom account team.
This section discusses various insights available with Quality Management.
The Interactions menu is the central location to review conversations and insights generated by Quality Management
Supervisors and administrators can centrally review all conversations analyzed by Quality Management through the Interactions sub-menu on the web portal. From this screen, an authorized user can click the blue Interaction ID in any row to open a detailed view of that Interaction. From there, the authorized user can play the recording, read the transcript, and view the associated analytics, comments, and details.
After selecting an Interaction ID, Quality Management’s conversational analysis is displayed on-screen.
The Analytics tab provides a conversational summary, including an engagement and sentiment score, next steps, Indicators and Topics identified, and more
When viewing an interaction, the Analytics tab provides a summary of conversational insights identified by Quality Management. These insights include:
Analytics also include an Agent Summary, providing details about how the agent interacts with the consumer, like the talk-to-listen ratio, filler words used, and more
Supervisors can use the Speaker Metrics report to see an overview of a team’s or individual agent’s performance over time
While supervisors can view an agent’s performance on a per-interaction basis, as seen above, the Speaker Metrics report helps supervisors see an overview of an individual agent’s performance over time. This can help supervisors track agent performance, including changes as a result of coaching and feedback, and identify trends over time, such as if an agent reduces (or increases) their speech events, talking speed, talk-to-listen ratio, or other Agent Summary metrics.
Each interaction includes a graph of the conversation’s sentiment over time, with links to the transcript and recorded sections for context
Interactions analyzed by Quality Management include a graph of the conversation’s sentiment over time. Each time a conversation’s sentiment changes, users can interact with the insight’s pop-up window, which includes a link to sections of the transcript for additional context, as well as an embedded play button to hear or watch that section of the conversation.
Sentiment score is currently determined through transcript analysis
At this time, Quality Management’s sentiment scores are calculated solely through AI analysis of a conversation’s transcript and do not account for other factors like tone, volume, or talk speed. For example, AI models may interpret a response of great as positive, while okay may be neutral, and whatever may be seen as negative in some contexts.
Indicators (Call Outs) are custom keywords or phrases that note critical moments in a conversation, like competitive brands, features, or products mentioned
Indicators, identified under the Call Outs section, are customizable keywords or phrases that are highlighted within a conversation’s analysis. Indicators can be used to capture critical moments of a conversation or track mentions of a specific competitor, feature, product, or phrase. Accounts can use indicators to identify specific elements of conversation that are worth reviewing or tracking.
Consider a business that is focused on reducing churn compared to competitors. The business can create Indicators for their products and those of their competitors. Consequently, if an Indicator is mentioned within a conversation, the business can quickly identify the context of the Indicator’s mention and proactively work to meet the consumer’s need or concern if one exists.
For example, if a business specializes in candy, they can create Indicators for the various types of candy, like Gummy Candy, along with specific examples of those candies, like gummy worms or gummy bears.
The following images provide an example of Indicators noted in a conversation’s analysis. Within this image, the Indicators Hard Candy and Gummy Candy are noted, because Indicator-specific keywords or phrases, like gummy bears and gummy worms were noted in the conversation’s transcript. After clicking an Indicator (i.e., gummy candy), the transcript will reflect sections in the conversation where the Indicators were noted.
Indicators are grouped under custom categories, with each category supporting multiple indicators
Accounts can create multiple, custom categories to group and track Indicators mentioned within conversation. This allows each account flexibility to create systems of information in ways that work best for their needs, environment, or sector.
For example, if a business specializes in chocolate candy, they could create a Competitors category, with an Indicator for each of their primary competitors’ candy types, like hard candy or gummy candy. Using Quality Management they could now create an Indicator for Hard Candy, with keywords for lollipops, candy canes, butterscotch, or other similar candies. Additionally, they could create a second Indicator for Gummy Candy, with keywords for gummy worms, gummy bears, fruit slices, or other soft candies.
When combined, account administrators can review which Indicators are most commonly mentioned, along with which of their competitors is mentioned, in addition to the specific product that is being mentioned. This can provide key context to a consumer’s sentiment and if they are interested in a competitor’s product(s).
The following image provides an example of the previous example’s Indicators and their associated keywords grouped under the common Competitors category.
Indicators match against the spoken language of an interaction
Quality Management will match Indicators against the spoken language of an interaction and its associated transcript. For this reason, Indicators should be developed in the language(s) your business expects to encounter.
Indicators can be retroactively applied to historical conversations
After adding new indicators to an account, interactions will retroactively reflect the new Indicators upon analysis, helping businesses quantify common themes over time as trends become clear.
Accounts are allowed up to 250 indicators
As of the date of this document’s publication, accounts are currently allowed up to 250 unique indicators.
Indicators can be set up to be triggered by agents, consumers, or both.
When creating an Indicator, account admins can configure if the Indicator will be noted in a conversation’s analysis based on who mentioned the keyword or phrase in the conversation.
For example, a contact center agent might regularly reference a product or set of features, which may skew analytics for Indicators mentioned. However, these data points may prove valuable if a consumer mentions these products or features, indicating their sentiment or business needs. Alternatively, an account may choose to always highlight indicator mentions, by either an agent or consumer, for quality assessment and complete coverage.
Supervisors and administrators can subscribe to Indicator mentions
Quality Management supervisors and administrators can subscribe to Indicator mentions via email to track or monitor specific themes or trends. When subscribing, the user can choose to receive subscription notifications on a daily, weekly, monthly, or continuous (i.e., as-it-happens), basis.
For instance, a contact center manager may subscribe to an Indicator for escalation requests that track the frequencies of consumers requesting to speak or escalate a situation to a manager.
Subscriptions can be specific to a keyword, phrase, or the entire Indicator
When subscribing, a user can choose to subscribe to the Indicator with all associated keywords and phrases, or, the user can subscribe to an individual or limited set of keywords.
For example, if the Indicator Cancel Service is selected, and contains the keywords cancel, terminate, discontinue, end contract, and termination, a user can subscribe to all keywords, or specific phrases that raise significant concern, like end contract.
The Indicator Mentions report provides insight into Indicator frequency and can help proactively identify trends
Account admins and supervisors can gain insight into Indicator frequency through the Indicator Mentions report. This dynamic report displays the frequency and percentage of interactions containing configured Indicators, phrases, or keywords, and can be used to view trends over time. This can help supervisors proactively monitor interactions and quickly identify issues and trends without manual searching, as seen in the following image.
After drilling into an Indicator’s analytics, additional information can be seen in line or bubble view for different perspectives of the same information. For example, the following image shows additional information for the Churn risk Indicator.
From there, supervisors can drill further into Indicator analysis and see specific conversations and moments in the transcript where the Indicator was mentioned. This includes the ability to filter data by account or user and drill down into specific keywords to see transcript snippets for context. For example, the following image shows mentions of the keyword Cancel, which is configured as a part of the Churn risk Indicator.
Topics track broad, predefined themes in a conversation and can provide insight to consumer concerns
Unlike Indicators, which focus on specific phrases or keywords, Topics highlight broader themes within a conversation. For instance, some subjects are discussed thematically from multiple perspectives, like a product’s security, privacy, or pricing. Instead of tracking these themes from a keyword-specific approach—as is done with Indicators—Quality Management uses artificial intelligence (AI) and machine learning (ML) algorithms to identify questions or statements that are relevant to a predefined theme.
For example, if a consumer is subject to an array of compliance or security regulations, they may commonly discuss or ask questions surrounding those topics. Consequently, when reviewing a conversation’s analysis within Quality Management, the analytics section may highlight portions of the conversation’s transcript that are related to those topics.
The following image provides an example of Topics noted under the Topic Highlights section. Within this image, the topics Variety, Quality, Regulation, Business Requirements, and Cost are highlighted, indicating that the conversation’s transcript included discussion around these topics.
Quality Management includes seven default topics and supports up to 10 additional custom topics
Quality Management includes seven default Topics it will automatically track within a conversation, including Pricing, Legal, Privacy, Security, User Requirements, Hardware, and Licensing. Account admins can additionally create up to 10 custom topics, but must provide a variety of guided sentences by which the Topic’s theme can be understood through AI and ML algorithms. The following image contains an example of the Privacy Topic and its foundational guiding sentences.
Topics match exclusively against the transcript of an interaction
Unlike Indicators, which identify spoken words within the spoken language, Quality Management matches Topics process through an AI/ML engine that currently exclusively supports the English language. Consequently, conversations in other support languages (e.g., Spanish) are first transcribed into English before matching Topics. When a Topic is created in a supported language other than English, it is translated to English for storage and used to match the translated transcript accordingly.
Speech Events note when an interaction had a period of silence or crosstalk above an account’s defined threshold(s), and if a call was put on hold
Speech Events help supervisors identify moments in an interaction where there was an extended moment of silence or crosstalk beyond an account’s defined threshold settings, in addition to the number of times a call was put on hold. This can help supervisors identify and confirm that interactions are meeting an organization’s standards, in addition to helping supervisors identify when in an interaction these events occurred.
Speech Events are outlined and labeled below the interaction’s media player
When viewing an interaction, supervisors can see an outline and chronology of the conversation’s Speech Events below the media player. This provides visual cues showing who spoke, when they spoke, and the type and duration of each event triggered, helping supervisors quickly identify these moments for review.
Quality Management Tools
This section discusses various tools available with Quality Management.
Performance Management provides features that assess the quality of your agent’s interactions and identify areas for improvement
Performance Management provides supervisors and administrators with the tools to systematically evaluate contact center interactions and identify areas for improvement through scorecards. Once the scorecard infrastructure is built within Quality Management, supervisors can conveniently review conversations either as part of a scheduled assessment or on a spontaneous (ad-hoc) basis, while also having the flexibility to assign multiple reviewers to the same interaction, helping ensure scoring consistency across a panel of assessors.
Scorecards are a list of user-defined questions used by a supervisor to score agent performance during consumer interactions
Within Quality Management, consumer interactions are evaluated through scorecards containing a list of user-defined questions intended to score an agent’s performance. Each scorecard template must be built by a Quality Management supervisor or administrator with sufficient role permissions before it can be applied to a conversation.
Scorecards support three question types—yes or no, single choice, and a rating scale—and can be used in any mixture
A Quality Management administrator may choose one of three question types for each question on a scorecard: yes or no, single choice, or a rating scale. Each scorecard may use any mixture of these question types as desired.
Scorecard questions support weighted values, cumulative scoring, evaluator comments, required responses, and automatic failures
Scorecard questions support a multitude of features to improve the evaluation process. Features available for scorecard questions include:
- Weighted Values: Questions within a section may contribute more or less points to a passing grade based on the importance of a question.
- Cumulative Scoring: Questions may be given a value which contributes to a cumulative evaluation score between all other scorecard questions. Depending on the agent’s actions, their cumulative score may or may not meet a passing grade by the end of an evaluation if they failed to follow proper procedure.
- Evaluator Comments: The scorecard evaluator may leave comments or notes for a specific question on the scorecard. This offers evaluators the opportunity to provide direct feedback on an agent’s performance, indicating what they have done well or where there is opportunity to improve.
- Required Responses: The scorecard question must be answered and cannot be left blank.
- Automatic Failure: If an agent fails a question, the entire scorecard is marked with a failing grade.
Each scorecard uses a percentage or points-based scoring system
When designing a scorecard, the Quality Management administrator may use either a percentage or points-based system. In a percentage-based system, the overall evaluation is expressed as a percentage (i.e. 90%), while in a points-based system, it is expressed as a numerical score (i.e., 90/100 or 133/150).
There is no significant difference between these two systems, except that a percentage-based scoring enables weighted scoring by section, whereas points-based scoring is exclusively cumulative.
Evaluations are when an assessor applies a scorecard to an agent’s conversation
An evaluation occurs when a Quality Management supervisor or manager applies a scorecard to an agent’s conversation. To perform an assessment, a user with sufficient role permissions can navigate to an interaction, select the Performance tab, click Evaluate, and select the appropriate scorecard.
Managers can assign evaluations to an assessor
In addition to spontaneous (ad-hoc) evaluations, a manager or user with sufficient role permissions can assign evaluations to other Quality Management supervisors or managers.
When an evaluation is assigned, the assessor will receive an email with a direct link to the conversation, the scorecard to be used, and a due-by date.
Agents can dispute or acknowledge their evaluations
Supervisors can require agents to acknowledge evaluations
If desired, account admins can require agents to acknowledge their completed evaluations. When this feature is enabled, agents receive email notifications for new evaluations that need acknowledgment. From there, agents can review and acknowledge or submit a dispute for their score. If an agent does not acknowledge or dispute the evaluation within the account’s set amount of time, it will be automatically acknowledged by the system. The following image provides an example of the email an agent will receive.
Calibrations allow multiple assessors to apply a common scorecard to the same conversation and are intended to ensure scorecards are being consistently applied
Customers can also add manual or “offline” interactions to score conversations not captured by their Contact Center integration
Supervisors and Managers, depending on their role’s permissions, can manually add “offline” interactions for scoring within Quality Management. This feature empowers these roles to apply scorecards to additional interactions not recorded by their Zoom Contact Center integration within a unified platform, helping to ensure a consistent set of criteria is used across all interactions.
Supervisors and Managers can leave comments on specific sections of a transcript for discussion, feedback, or follow up
Supervisors and managers can leave comments at specific timestamps within each interaction analyzed by Quality Management. This feature allows them to provide direct feedback to agents on critical moments of the conversation or mark sections for future discussion or follow-up. Additionally, commenters can tag an agent or supervisor, who will receive a notification and bring them to the comment. These comments can be left as public for anyone to see, or private for only tagged persons.
The following image provides an example of a comment left within a conversation.
Moments are short, recorded segments — or clips — of an interaction and can be shared with others
Within Quality Management, Moments are short recordings that highlight important moments from a conversation and can be shared with others for viewing or learning opportunities.
For example, if an agent handles a difficult situation or moment in a conversation remarkably well, a supervisor or manager can create a separate, shorter, recording of the conversation highlighting what the agent did well. Afterwards, the supervisor or manager can share the recording link of the Moment with others—internal or external to the account—as an example for other agents.
Account Management
Account Settings
Quality Management must be enabled on the account level to access recordings and transcripts for analysis
Quality Management requires access to an account’s Zoom Contact Center recordings and transcriptions to generate insights and analyses, and must be enabled within the Zoom web portal by an account administrator or authorized user. To grant access, perform the following steps:
1. As an account owner, or an account authorized to edit account settings, sign in to the web portal.
2. Under the Account Management sub-menu, click Account Settings.
3. Click the Quality Management tab along the above the settings list. Be aware you may need to click < > arrows to scroll to the option.
4. Enable the setting to Analyze Zoom Contact Center Recordings and Transcripts
Account administrators can further customize Topics, Indicators, and thresholds for silence and crosstalk events
Account administrators or authorized users with sufficient general (non-Quality Management) role-based permissions can edit and customize settings for Topics, Indicators, and thresholds for silence and crosstalk events. To access the menu for these settings, perform the following steps:
1. As an account owner, or an account authorized to edit account settings, sign into the web portal.
2. Under the Account Management sub-menu, click Account Settings.
3. Click the Quality Management tab along the above the settings list. Be aware you may need to click < > arrows to scroll to the option.
4. Interact with the appropriate setting.
User Licensing and Roles
Quality Management utilizes a per-seat license model
Agents and supervisors are granted access to Quality Management through a per-seat licensing model and each user that requires access to any features must be assigned a license for the product. For example, one supervisor and nine agents would require a total of 10 Quality Management licenses.
Users must be assigned a Quality Management license to access and use the product
User accounts must be assigned a Quality Management license (either a-la-carte or through a bundled license) by an account admin or authorized user to access and use the product.
Quality Management provide four different role types with customizable permissions in addition to custom role support
Quality Management includes four different role types, granting users varying levels of access based on business needs. Although Zoom provides default permissions for each role, account admins can create custom roles or customize the default roles based on unique team dynamics. The default roles are listed in the following table:
Role |
Description |
|---|---|
Quality Management Admin |
Administrators have a wide range of permissions for accessing and managing Zoom Quality Management for your account. |
Quality Management Supervisor |
Supervisors manage a group of agents and have access to review their interactions and evaluations. |
Quality Management Agent |
Agents have access to limited Zoom Quality Management functionality but no account management privileges. This is the default role upon license assignment. |
Quality Management Evaluator |
Evaluators have access to review interactions and assess performance using scorecards |
To update or create a custom role with applicable permissions, perform the following steps:
- As a Zoom account owner or admin authorized to edit roles, navigate to the Role Management page on the web portal.
- Click the Quality Management tab.
3. Click the pencil icon to the right of the role you are editing or click + Add Role.
4. Define the role access as desired.
5. Repeat for any additional roles.
Quality Management Roles are unique to the service and do not override or affect other user roles
User roles specific to Quality Management are scoped exclusively to the service and do not conflict with other user roles on the account. Consequently, a user can be a general account admin without Quality Management admin permissions and vice versa.
Admins can assign Quality Management licenses through SAML response mapping, SCIM, or manual assignment
Account admins can manage Quality Management license assignments via single sign-on (SSO) with SAML response mapping or SCIM, or manually through the web portal.
Data Privacy & Security
Conversations analyzed by Quality Management are not used to further train Zoom’s or third-party AI models
Zoom does not use any of your audio, video, chat, screen sharing, attachments, or other communications-like Customer Content (such as poll results, whiteboard and reactions) to train Zoom’s or its third-party artificial intelligence models.
Quality Management cloud recording files are subject to the same retention policies as other Zoom Contact Center
Quality Management interaction recordings are subject to the same retention policies as other Zoom Contact Center recordings. Accordingly, interactions within Quality Management will be deleted when the recording file is deleted within Zoom Contact Center.
Recording disclaimers and consent prompts are managed through Zoom Contact Center
All recording consent prompts and disclaimers are managed through Zoom Contact Center’s settings. Quality Management does not provide unique prompts related to the service.
Quality Management is available globally from Zoom’s U.S.-based cluster and locally within Zoom’s European-based cluster
Quality Management is globally available to Zoom customers hosted on Zoom’s United States-based cluster and regionally within Zoom’s European-based cluster. Quality Management customers currently cannot configure Quality Management-specific data for storage outside of their hosted account’s region.
All data generated through use of the service is encrypted at rest and stored in AWS
Quality Management account data is encrypted at rest with a 256-bit AES-GCM algorithm, using encryption keys stored in Amazon Web Services Key Management Service (AWS KMS).
